A founder sent us an AEO consultant's proposal in August. Twelve pages, a clean deliverables list, $4,000 a month, and a line near the end promising weekly citation reporting across ChatGPT, Perplexity, Gemini, AI Overviews and AI Mode.
We asked one question back: how many prompts, and how many hours a month does the retainer include.
Twelve prompts. Ten hours.
Nothing in that proposal was dishonest. The scope was real and the person was good. The arithmetic just did not close, and no page on the first search results page for this query does the arithmetic at all.
What does an AEO consultant do?
An AEO consultant is one person you hire to get your brand named inside AI answers. They audit which engines cite you against named competitors, rewrite pages into passages a model can lift, fix entity and schema gaps, and tell you where your category's citations actually come from. The measurement usually runs on a platform seat, not by hand.
The capacity ledger
What a one-person AEO retainer can hold, in hours
Layer one is what reading the measurement costs. Layer two is what the retainer contains. Layer three is the price of not doing it by hand.
Layer 1: monthly reading load, five engines, one weekly pass, one minute per answer
10 prompts · 217 answers/mo
3.6 hours/mo to read
Affordable on any retainer. Otterly measured this panel reporting anywhere from 39% to 71% for a brand whose true coverage is 55%, so the hours buy a number nobody can act on.
50 prompts · 1,083 answers/mo
18.1 hours/mo to read
The knee of the precision curve and the floor worth insisting on. Also more hours than the entry retainer contains.
100 prompts · 2,167 answers/mo
36.1 hours/mo to read
Nine points of band width. Reading it by hand is most of a full-time fortnight every month.
Layer 2: hours a retainer contains at $300 an hour, against the 18.1 hours a 50-prompt panel needs
The filled portion of each bar is measurement, not work
At the entry retainer the reading does not fit at all. At $6,000 it takes 90% of the month and leaves under two hours for the rewriting, the schema and the off-domain placement you are paying for.
Layer 3: published monthly seat prices for the instrument that removes those 18 hours
Three of those four cost less than one billed hour at $300, and each replaces eighteen. A consultant who is still reading answers by hand is charging you consulting rates for data entry.
That last sentence is the whole post. Everything a consultant sells is bounded by how many hours they have, and the measurement half of answer engine optimization has a fixed hourly cost that nobody quotes.
Scope is what a consultant writes down. Hours are what they actually have.
6 reasons a solo AEO consultant retainer runs out of hours
These are structural. They apply to a good consultant and a bad one identically, which is why they are worth understanding before you evaluate any particular person.
Reason #1: A panel small enough to read by hand is too small to read at all
Otterly published the measurement in September 2026 that this category had been missing: 252,407 AI search answers across seven engines, 520 prompts, windows of 40 to 91 days.
The result is a precision curve. At 10 tracked prompts, 90% of brand coverage figures fell inside a band 32 points wide, running from 39% to 71%.
At 50 prompts the band narrows to 13 points. At 100 prompts, 9 points.
A twelve-prompt panel is cheap to read and cannot resolve anything. That is the trap the proposal above walked into.
Reason #2: The reading itself costs about eighteen hours a month
Take the 50-prompt floor and five engines a B2B buyer cares about. One weekly pass is 250 answers, or roughly 1,083 a month.
Each answer has to be classified four ways: were you mentioned, were you cited with a link, were you recommended as the pick, and which sources did the model lean on. A minute each is generous rather than slow.
That is 18.1 hours a month, spent before a single page is rewritten.
Reason #3: The entry retainer does not contain eighteen hours
OuterBox's August 2026 rate guide puts an independent marketing consultant at $75 to $250 an hour and a senior or fractional specialist at $150 to $500. Call it $300 for someone who can actually do this work.
At $300, a $3,000 retainer is ten hours. A $6,000 retainer is twenty. The reading needs eighteen.
So the entry engagement cannot fit the measurement at all, and the next one up spends 90% of the month on it.
Reason #4: One cadence cannot serve engines that behave nothing alike
The same Otterly study measured how much of what an engine cited yesterday it cites again today. Perplexity reused about 75% of its sources day to day. ChatGPT reused about 26%.
A weekly check reports a durable position on one engine and close to a coin flip on the other. Both land in the same column of the same deck.
Fixing that means running different cadences per engine, which multiplies the hours in Reason #2 rather than reducing them.
Reason #5: A single reading is not a result, and convergence takes many
Sielinski's convergence work, a July 2026 preprint, found that rankings needed between 33 and 94 answers to settle across 30 platform-topic combinations. Three combinations never settled.
Read that against a once-weekly, one-run-per-prompt schedule. Four readings a month does not approach 33, so the monthly number is a snapshot of noise presented as a trend.
Reason #6: The larger half of the job is not on your website
Our concluded CITE Index study ran 500 buyer prompts nightly through ChatGPT, Gemini and Google AI Mode for 63 days, collecting 90,132 answers across ten categories.
Four of the twelve most-cited domains were brand-owned. The other eight were places you have to earn your way into.
Reddit alone drew 14,698 citations, appearing in 13.6% of all answers. The full corpus sits in the final report.
Most of your citation pool sits on domains you cannot edit. Relationship work does not compress into a 10-hour month.
What a consultant's page promises:
- •Full-funnel AEO strategy
- •Coverage across every major AI engine
- •Weekly citation reporting
- •Content, schema and entity optimization
- •Off-page authority building
What ten hours a month physically holds:
- •One narrow panel, read once
- •Two or three pages rebuilt into answer blocks
- •One schema pass
- •A call
- •Nothing on the eight domains you do not own
Both lists can be true about the same person. Only the second one describes what you receive.
What an AEO consultant costs in 2026
MarketerHire's hiring guide publishes the clearest bands on the first results page, and they match what we see in client procurement. What it does not publish is what each band buys in hours, so that column is ours.
| Engagement | Published range | Hours at $300 | What it realistically buys |
|---|---|---|---|
| Fractional retainer, entry | $3,000 a month | 10 | Advisory and page repair only. The measurement has to come from your seat or not happen. |
| Fractional retainer, working | $6,000 to $8,000 a month | 20 to 27 | A real panel on a platform seat, plus four to six pages rebuilt a month. |
| Fractional retainer, upper | $10,000 to $15,000 a month | 33 to 50 | The above plus sustained off-domain placement, which is the half most retainers skip. |
| Fixed-scope project | $8,000 to $25,000 | 27 to 83 | A diagnosis you keep. The best shape for a first engagement. |
| Full-time hire | $140,000 to $220,000 base | n/a | Only rational once the weekly loop is already running and someone has to own it. |
MarketerHire also states that citation lift typically appears in 60 to 90 days and share-of-voice moves over two to three quarters. Those numbers are unsourced on the page, and our own corpus suggests the second half is closer to right than the first. Across 63 days the category leader changed on only 18.7% of day pairs, and in four of ten categories it never changed once.
Getting into the answer and taking the top slot are different purchases at different prices.
Find out whether your gap is absence or displacement
We run your buyer prompts to convergence across every major AI engine, map the source pool your category actually cites, and hand you the panel and the baseline. You keep both whether or not you work with us after.
Book a Discovery CallHow to hire an AEO consultant without buying a guess
The vetting checklists on this query all ask about process. Process questions cannot be failed. These five can.
Step 1: Ask which platform seat runs the measurement and whose name is on it
There are three possible answers. They run a seat you pay for, they run their own and pass the data through, or they read answers by hand.
Published seat prices run from $99 a month at Rankscale Pro to $189 for Otterly Standard, $295 for AthenaHQ Starter and $415 for Peec across five engines. Three of those cost less than one billed hour.
A consultant still reading by hand in 2026 is charging consulting rates for data entry, and you are funding it out of the same ten hours.
Step 2: Ask for the prompt count, the engine list and the runs per prompt in writing
Not "coverage across every major engine." Three integers and a cadence.
Fifty prompts is the floor Otterly's curve supports. Below that, insist the proposal says so, because a small panel is a legitimate choice at a legitimate price and only becomes a problem when it is reported as precision.
Step 3: Add ten prompts you wrote yourself before the baseline is set
This is the single most informative thing you can do, and it costs nothing.
The score is a function of the prompt set, and the prompt set is normally chosen by the person being graded on it. A preprint submitted in September 2026 by Olivier Martinez makes the formal version of this argument: a prompt corpus defines an answer market that need not represent real demand.
Watch what your ten prompts do to the baseline. That movement is the honest number.
Step 4: Require mentions, citations and recommendations to be counted separately
Being named in prose, being cited as a linked source, and being recommended as the pick are three commercial events worth very different amounts.
Blended into one "AI visibility" score, a provider can raise the headline by getting you mentioned more often in answers that still recommend a competitor. It is the easiest number in this category to move without moving anything you care about.
Step 5: Split the first engagement into a project, then decide on the retainer
Buy the diagnosis as fixed scope: the panel, the baseline, the source-pool map, the page-level gaps. You keep the artifacts regardless of what happens next.
Then you are choosing a retainer with a number in front of you rather than a deck. We wrote up how the same decision plays against internal capacity in GEO in-house versus agency.
When a consultant is the wrong shape for the problem
A consultant is the right call when the problem is diagnosis, page-level repair, or a team that knows what to do and needs someone to tell them what is actually happening. One good operator beats a junior pod for that work every time.
A consultant is the wrong call when the eight domains you do not own are the binding constraint. Earning placement in the review sites, forums and publications your category's answers lean on is relationship work with a long clock, and it does not compress into ten hours a month no matter who holds them.
That is the split to make before you compare rates. If you want the whole loop run rather than advised on, a managed AEO service carries the measurement cost as overhead instead of billing it against your hours, and we laid out how to read that market in how to choose an AEO agency.
The failure mode we see most often is buying a consultant for an agency-shaped problem, then concluding after two quarters that AEO does not work.
FAQ
How much does an AEO consultant cost?
Fractional retainers run $3,000 to $15,000 a month, fixed-scope projects $8,000 to $25,000, and full-time hires $140,000 to $220,000 base. At a typical $300 hourly rate the entry retainer is ten hours, which is less than the reading load of a defensible measurement panel.
Is AEO consulting different from SEO consulting?
Yes, in what gets optimized. SEO work targets a ranked position for a page. AEO work targets extraction of a passage and citation of a source, across five engines that disagree with each other, and much of it happens on domains you do not own.
How long does an AEO consultant take to show results?
Published guides say citation lift in 60 to 90 days and share of voice over two to three quarters. Our 63-day corpus supports the slower half: the category leader changed on only 18.7% of day pairs, and never at all in four of ten categories. Appearing in answers is faster than displacing whoever owns them.
Should I hire an AEO consultant or an AEO agency?
Hire a consultant for diagnosis, advisory and page-level repair. Hire an agency when the weekly measurement loop and off-domain placement both need running, because those two costs are overhead an agency spreads and a solo retainer bills you for directly.
Do I need an AEO consultant if I already have an SEO agency?
Usually you need the measurement first, not another provider. Run one panel, see whether your gap is absence from the answers or displacement inside them, and then decide whether your current agency can be pointed at it. We cover the diagnostic in what an AI visibility audit should tell you.
What to do with this
Ask any consultant you are considering for three integers: prompts, engines, runs per prompt. Then divide the retainer by $300 and see whether the hours cover the reading.
If they do not, the engagement is advisory. Advisory is worth buying. It is just worth buying at the advisory price, with the measurement running somewhere you control.
Get the baseline before you sign anything
We build the prompt panel, run it to convergence across ChatGPT, Gemini, AI Mode, AI Overviews and Perplexity, and give you the source-pool map for your category. Yours to keep, and usable with any provider you pick.
Book a Discovery CallContinue the brief
How Do You Choose an AEO Agency in 2026?
Every AEO agency shortlist on page one was written by an AEO agency that ranked itself first. Here is what to check instead, and what the work can move.
Do AI Search Engine Optimization Services Work?
AI search engine optimization services can work. Most reported wins are sampling noise. Here is the buyer test that tells the two apart before you sign.
What Does a Mature AEO Program Look Like in 2026?
Conductor surveyed 250+ enterprise marketing leaders. 51% run AEO on integrated platforms, and high-maturity teams are 6x more likely to. Here is the bar.
Framework
Learn the CITE framework behind our GEO and AEO work
See how Comprehend, Influence, Track, and Evolve turn AI visibility into an operating system.
Services
Explore our managed GEO services and AEO execution model
Audit, prompt discovery, content execution, and ongoing monitoring tied to AI search outcomes.
Audit
Start with an AI visibility audit before execution
Understand prompt coverage, recommendation gaps, source mix, and where competitors are winning.
